[Collins] 32S-1 Oscillations & Other Troubles

[Dr. Gerald N. Johnson, electrical engineer]

Henry Turner wrote:
> 
> Hello Jerry,
> 
> I was wondering, for the benefit of others on the list
> if you could describe in detail what are the symptoms of a
> Collins rig with "bad" neutralization.  When it is necessary
> and just what happens if it is not done.  How can one tell.
> 
> I always enjoy when you post... I learn something every day.
> 
> Henry
> K5YDD
> 

Neutralization the name, comes from neutralizing the feed back
capacitance (plate to grid) of the tube. Could be called
unilateralization, to be sure the coupling from output to input is
smaller than the gain from input to output. If neutralization isn't
performed the main symptom is stage oscillation, more likely on the
higher bands than the lower bands because the coupling of a series
capacitor is greater at high frequency than at low frequency.

The strongest symptom of imperfect neutralization is oscillation. Its
possible that positive feedback could cause a stage to distort from that
feedback. Normally a linear stage has more stage gain than a Class C
stage using the same tube so keeping it linear demands better
neutralization. A stage with slightly defective neutralization may
perform OK, but will hint at that imperfection by the output power peak
not coinciding precisely with the plate current dip. That's because the
feedback is affecting the stage gain and the RF voltage on the plate is
greatest when the plate tank is resonant, but the feedback phase is
affected by the phase angle of the plate tank (resistive when resonant,
capacitive or inductive when not resonant) and that signal coupled to
the grid changes the effective gain of the stage according to the plate
tuning.

In simpler terms, with proper neutralization the stage doesn't oscillate
at the signal frequency or the band its tuned to and the power output
peak matches the plate current dip. Imperfect neutralization tends to
make the plate current dip and peak output not match when tuning. Bad
neutralization lets the stage oscillate at the tuned frequency.

Neutralization is always required when tubes are changed because the
required precision of neutralization is more critical than the variation
in tube characteristics between tubes of the same type. A, B, and W
suffixes of 6146 represent more than production variations in tube
characteristics with B (and some W) being enough different the 32S-1
neutralization circuit lacks the adjustment range to achieve
neutralization.

Glen Zook says the 32S-1 neutralization circuit will "burn up" if it
happens to neutralize the 6146B. That's understandable because the
greater capacitance of the 6146B demands greater capacitance in the
neutralization circuit which leads to greater RF current and that heats
the components at the rate of the square of the current. So a 20% rise
in current results in a 44% rise in heat. Smoke indeed.

73, Jerry, K0CQ, Technical Advisor to the CRA
-- 
Entire content copyright Dr. Gerald N. Johnson, electrical engineer.
Reproduction by permission only.